CN115378812A - Method and system for data center network equipment maintenance - Google Patents

Abstract

A method and system for maintaining network devices in a data center involves a network upgrade tool coupled to a CMDB, a rules DB, an upgrade path DB, and a network configurator. The network upgrade tool is configured to collect/obtain information from the network devices, remove those non-compliant network devices from automated maintenance, cluster the remaining network devices into groups of redundant network devices, and for all such groups, upgrade the groups in turn The network operating system of the network device present in .

Description

Method and system for data center network equipment maintenance

technical field

The technology relates to information technology, and more particularly to methods and systems for automated maintenance of network equipment in data centers.

Background technique

Systems have been developed aimed at supporting the maintenance of network equipment such as switches, routers, etc. in data centers. As used herein, "maintenance" may include, for example, performing an upgrade of a device from a current network operating system to a target (eg, latest) network operating system. As used herein, a "data center" is not limited to infrastructure located within the physical boundaries of a server farm, but includes infrastructure deployed, controlled and maintained by an organization for either the organization's own internal services or to third party entities, i.e. all of the organization's customers All infrastructure that provides services based on computer clusters, whether local or remote. There has been a long-standing need for a system that automates maintenance of network equipment in a data center while saving required manpower, avoiding human error, and managing customer impact.

However, the challenge in developing such systems lies not only in the number of network devices to address, but also in the diversity of such devices in the data center, including their characteristics and their roles. For example, a data center network may consist of different architectures that work side-by-side to provide connectivity and services, where each of these architectures is designed with specific features, hardware models, enabling characteristics, and/or levels of redundancy And designed for that specific feature, hardware model, enabling feature, and/or level of redundancy, and all of these parallel fabrics are connected to a more central fabric or core network.

The challenge of such a development is also to minimize disruption to the functionality of the data center and to ensure that the unavailability of some or all of the data center due to maintenance of network equipment is limited as much as possible. This in turn allows minimizing disruptions in network connectivity services provided by the organization to customers, thereby meeting contractual service quality commitments as the case may be.

Therefore, there is a need for a method and system that allows a simple entry point for data center operators without specific network knowledge by providing a list of network devices to be upgraded and some simple options to enter if and when needed To automate large-scale network equipment maintenance activities.

In general, the present technology aims to make data centers more efficient by removing from automated maintenance those network devices that are at higher risk of automated maintenance failure and utilizing the redundancy of the remaining network devices in order to limit downtime in the data center due to maintenance operations. Maintenance automation of network devices in .

Subject matter discussed in the Background section should not be admitted to be prior art solely because it is mentioned in the Background section. Similarly, it should not be assumed that a problem mentioned in or associated with the subject matter of the Background section had been previously recognized in the prior art. The topics in the background section merely represent different approaches.

Contents of the invention

Embodiments of the present technology were developed based on the developers' recognition of shortcomings associated with the prior art.

In one embodiment, various embodiments of the present technology provide a method for maintaining network equipment in a data center, the method comprising:

- Collect for each of the network devices on the maintenance list:

- Management IP (MANAGEMENTIP) information, where MANAGEMENTIP represents a virtual identity uniquely associated with a network device; and

- any label and location (LOCATION) information, where the label represents a key-value tuple associated with the network device, where the key is any of BU, ROLE, or INFRA, and the value is a pointer to the actual data or data for the key, Among them, the BU value indicates the characteristics of the network equipment related to the service or product supply used by the network equipment in the data center, the ROLE value indicates the characteristics of the network equipment related to the position and function occupied by the network equipment in the data center, and INFRA the value represents a characteristic of the network device related to the version or generation of data center infrastructure in which the network device is operating, and wherein the LOCATION information represents a characteristic of the network device related to the actual physical location of the network device;

- For each network device in the network devices on the maintenance list, use the corresponding MANAGEMENTIP to obtain the hardware model and current network operating system level of the network device;

- Remove network devices with unsupported hardware models or current network operating system levels or errors in any of the BU, ROLE or INFRA labels or LOCATION information from the maintenance list;

- Group the remaining network devices on the maintenance list into: i clusters BU _i of network devices with the same BU label, j cluster ROLE _j of network devices with the same ROLE label in the BU _i cluster, and the same INFRA k clusters Cluster _ijk in the ROLE _j cluster of the network device of the label;

- Create redundant groups of network devices within each Cluster _ijk cluster according to the redundancy rules applicable to INFRA _k ;

- Verify in each created group that the size rules for the combination BU _i -ROLE _j -INFRA _k applicable to the number of network devices in the group are complied with and that all network devices in the group share the same LOCATION information;

- Remove from the maintenance list network devices that do not have a cluster or grouping group, or are present in a group that does not comply with redundancy rules or size rules, or do not share the same LOCATION information when present in the same group; and

- Using the corresponding MANAGEMENTIP and upgrade rules, for all groups in all Cluster _ijk clusters, upgrade the remaining network devices in each created group from the current version of the network operating system to the target version of the operating system in sequence.

In an embodiment, the collection and acquisition are performed using a unique ID assigned to each of the network devices on the maintenance list by the data center operator when deploying the data center, and the unique ID is associated with each network device Associate any of the tags in the , along with the MANAGEMENTIP and LOCATION information.

In an embodiment, removing from the maintenance list also includes removing network devices that do not have exactly one value for each of the BU, ROLE, and INFRA tags.

In an embodiment, removing from the maintenance list further includes creating an error list populated with IDs of removed network devices.

In an embodiment, the error list also includes a sub-list of errors that can be corrected by the data center operator and errors linked to faulty network equipment.

In an embodiment, the sequential upgrading further includes adjusting the number of parallel processing Cluster _ijk clusters to minimize the downtime of the data center.

In an embodiment, sequentially upgrading further includes adjusting the order of processing the Cluster _ijk clusters to optimize continuity of service to customers of the organization that deploys, controls and maintains the data center.

In yet another embodiment, various embodiments of the present technology provide a system for maintaining network devices in a data center that includes a network upgrade coupled to a CMDB, a rules DB, an upgrade path DB, and a network configurator tool, the network upgrade tool is configured to:

- Collect from the CMDB for each of the network devices on the maintenance list:

- MANAGEMENTIP information, where MANAGEMENTIP represents a virtual identity uniquely associated with a network device, and

- Any label and LOCATION information, where the label represents a key-value tuple associated with a network device, where the key is any of BU, ROLE, or INFRA, and the value is a pointer to the actual data or data for the key, where BU represents The characteristics of the network equipment related to the service or product supply that the network equipment is used in the data center, ROLE indicates the characteristics of the network equipment related to the position and function that the network equipment occupies in the data center, and INFRA indicates the network equipment and A version or generation-related characteristic of the data center infrastructure in which the network device operates, and wherein LOCATION represents a characteristic of the network device that is related to the actual physical location of the network device;

- For each network device in the network devices on the maintenance list, use the corresponding MANAGEMENTIP to obtain the hardware model and current network operating system level of the network device;

- Remove network devices with unsupported hardware models or current network operating system levels or errors in any of the BU, ROLE or INFRA labels or LOCATION information from the maintenance list;

- Group the remaining network devices on the maintenance list into: i clusters BU _i of network devices with the same BU label, j cluster ROLE _j of network devices with the same ROLE label within the BU _i cluster, and the same INFRA label k clusters Cluster _ijk in the ROLE _j cluster of network equipment;

- Create redundant groups of network devices within each Cluster _ijk cluster according to the redundancy rules applicable to INFRA _k collected from the rules DB;

- In each created group verify compliance with the size rules collected from the rules DB for the combination BU _i -ROLE _j -INFRA _k applicable to the number of network devices in the group and that all network devices in the group share the same LOCATION information;

- Remove from the maintenance list network devices that do not have a cluster or grouping group, or are present in a group that does not comply with redundancy rules or size rules, or do not share the same LOCATION information when present in the same group; and

- Use the corresponding MANAGEMENTIP and the upgrade rules collected from the rule DB to upgrade the remaining network devices in each created group sequentially from the current version of the network operating system for all groups in all Cluster _ijk clusters through the network configurator to the target version of the operating system.

In an embodiment, the network upgrade tool is further configured to collect and acquire using a unique ID assigned to each of the network devices on the maintenance list by the data center operator when deploying the data center, and the unique ID Associated with either of the tags for each network device, along with the MANAGEMENTIP and LOCATION information.

In an embodiment, the network upgrade tool is further configured to create an error list populated with IDs of network devices removed from the maintenance list.

In an embodiment, the error list also includes a sub-list of errors that can be corrected by the data center operator and errors linked to faulty network equipment.

In an embodiment, the network upgrade tool is further configured to adjust the number of parallel processing Cluster _ijk clusters to minimize data center downtime.

In an embodiment, the network upgrade tool is further configured to adjust the order of processing the Cluster _ijk clusters to optimize continuity of service to customers of the organization that deploys, controls, and maintains the data center.

In an embodiment, redundancy rules and size rules are hardcoded in the network upgrade tool.

In yet another embodiment, various implementations of the present technology provide a computer-readable medium including instructions for causing a computing system to perform the method described above.

In yet another embodiment, various embodiments of the present technology provide a method for maintaining a plurality of network devices in a data center. The method includes collecting for at least one network device of a plurality of network devices, the at least one network device selected from a maintenance list: - MANAGEMENTIP, which is a virtual identity uniquely associated with the at least one network device; - LOCATION, the LOCATION is information representing the actual physical location of at least one network device; - a label, which is a key-value tuple associated with at least one network device, the key value being one of data and a pointer to data,

The label is one of the following: - BU label, which is information indicating the product supply application of at least one network device in the data center; - ROLE label, which indicates that at least one network device in the data center information about the function occupied; and - the INFRA tag, the INFRA tag is information indicating the version of the data center in which the at least one network device is running; for at least one network device on the maintenance list, based on the MANAGEMENTIP of the at least one network device, obtaining a hardware model and a current network operating system level of at least one network device; and in response to determining at least one of: the hardware model of at least one network device is not supported, the current network operating system level of at least one network device is not supported Support, errors in at least one of the BU label, ROLE label, INFRA label and LOCATION label, remove at least one network device from the maintenance list.

In some embodiments, the method also includes determining at least one of the following: the hardware model of at least one network device is not supported, the current network operating system level of at least one network device is not supported, BU label, ROLE label, Error in at least one of the INFRA tag and the LOCATION tag.

In some embodiments, the method further includes collecting, for each network device in the plurality of network devices: MANAGEMENTIP, LOCATION, and label; in response to removing at least one network device from the maintenance list, maintaining the remaining network devices on the list Clustering into: i cluster BU _i of network devices with the same BU tag value in one of the associated tags of the remaining network devices; j of network devices with the same ROLE key value in one of the associated tags of the remaining network devices cluster ROLE _j , and k clusters Cluster _ijk within the ROLE _j cluster of network devices having the same INFRA tag value in one of the associated tags of the remaining network devices; create redundant multiples within each Cluster _ijk cluster groups of network devices; verify that _the number of redundant network devices in each created _group matches the first number according to the redundancy rules applicable to INFRA _k ; and verify that _the number of redundant network devices in each created group matches the first number; and , verifying that the number of network devices in each group matches the second number.

In some embodiments, the method further includes verifying whether the network devices of each group share the same LOCATION; At least one network device present in the given group whose number does not match the second number, or present in another group and not sharing the same LOCATION.

In some embodiments, the method further includes upgrading the remaining network devices in each group within all Cluster _ijk clusters from the current version of the network operating system to the target version of the operating system using the upgrade rule and the corresponding MANAGEMENTIP.

In some embodiments, the collection and acquisition is performed using a unique ID assigned to each of the network devices on the maintenance list by the data center operator when the data center is deployed, and the unique ID is associated with each network Any of the device's associated tags, MANAGEMENTIP, and LOCATION are associated.

In some embodiments, removing from the maintenance list also includes removing network devices that do not have exactly one value for each of the BU, ROLE, and INFRA tags.

In some embodiments, removing from the maintenance list further includes creating an error list populated with IDs of removed network devices.

In some embodiments, the error list also includes a sub-list of errors that can be corrected by the data center operator and errors linked to faulty network equipment.

In some embodiments, upgrading also includes adjusting the number of parallel processing Cluster _ijk clusters to minimize data center downtime.

In some embodiments, upgrading also includes adjusting the order of processing the Cluster _ijk clusters to optimize continuity of service to customers of the organization that deploys, controls, and maintains the data center.

In yet another embodiment, various embodiments of the present technology provide a system for maintaining network devices in a data center that includes a network upgrade coupled to a CMDB, a rules DB, an upgrade path DB, and a network configurator tool, the network upgrade tool is configured to: - collect from the CMDB for each network device in a plurality of network devices on the maintenance list: - MANAGEMENTIP, which is a virtual identifier uniquely associated with a given network device;- LOCATION, which is information representing the actual physical location of a given network device; and - a label, which is a key-value tuple associated with the given network device, wherein the label is one of:- - information indicating the business or product supply that the network equipment is used for in the data center, BU; - information indicating the position and function occupied by the network equipment in the data center, ROLE; and - indicating the data center in which the network equipment operates Version or generation information, INFRA; the value used for the label is the actual data or a pointer to the data; - For each network device in the network device on the maintenance list, use the corresponding MANAGEMENTIP to obtain the hardware model and CURRENT NOS LEVEL; - Removes from the maintenance list that has an unsupported hardware model or current NOOS level either in any of its corresponding associated BU, ROLE, and INFRA tags or in its LOCATION There is a wrong network device in any LOCATION in; - cluster the remaining network devices on the maintenance list into: i cluster BU _i of network devices with the same BU key value in one of the associated labels of the remaining network devices, j cluster ROLE _j of network devices with the same ROLE key value in one of the associated labels of the remaining network devices within the cluster of BU _i , and the network with the same INFRA key value in one of the remaining network devices' associated labels k clusters Cluster _ijk within the ROLE _j cluster of the device; - according to the redundancy rules applicable to INFRA _k collected from the rule DB, create redundant groups of network devices within each Cluster _ijk cluster; - according to the applicable Redundancy rules for INFRA _k , verify that the number of redundant network devices in each group matches the first number; - verify the number of redundant network devices in each group according to the size rules applicable to the combination BU _i -ROLE _j -INFRA _k whether the number of network devices matches the second number; - verify that all network devices in each created group share the same LOCATION; - remove from the maintenance list the number that exists in one of the groups and is redundant with Network devices that do not match in the first quantity, are present in one of the groups in a quantity that does not match the second quantity, or are present in one of the groups and do not share the same LOCATION; and - via the Network Configurator, Use upgrade path D from The upgrade rules and corresponding MANAGEMENTIP collected in B upgrade the remaining network devices in each group from the current version of the network operating system to the target version of the operating system for all groups in all Cluster _ijk clusters.

In some embodiments, the network upgrade tool is further configured to collect and obtain using a unique ID assigned to each of the plurality of network devices on the maintenance list by the data center operator when deploying the data center, and This unique ID is associated with any one of the associated tags of each network device, the MANAGEMENTIP and the LOCATION.

In some embodiments, redundancy and size rules are hardcoded in the network upgrade tool.

The technology provides a method and system that will perform exactly the same maintenance when network devices conform to the same parameters (architecture, role, hardware model, running software version and/or enabled features, etc.) process, thereby removing the risk of human error. The adjusted amount of parallel execution will limit the impact of data center unavailability. Network connectivity services provided to customers will be enhanced by identifying redundancies in the infrastructure of network equipment sharing certain parameters or characteristics and non-simultaneous maintenance of such network equipment so as to potentially degrade the service but not completely interrupt it interruptions are minimized. Furthermore, such methods and systems would allow data center operators to easily consider new equipment hardware or new architectures between equipment.

In the context of this specification, unless expressly stated otherwise, a system may refer to, but is not limited to, an "electronic device", "operating system", "computing system", "computer-based system", "controller unit", "monitoring device", "control device" and/or any combination thereof.

In the context of this description, the functional steps shown in the figures may be provided by the use of dedicated hardware as well as hardware capable of executing software in association with appropriate software. Furthermore, the various functional blocks shown in the figures, such as the functionality labeled "network device", "tool", "configurator", etc., may be provided through the use of dedicated hardware as well as hardware capable of executing software associated with appropriate software . When provided by a "processor," the functionality may be provided by a single dedicated processor, a single shared processor, or multiple separate processors, some of which may be shared. In some embodiments of the present technology, the processor may be a general-purpose processor, such as a central processing unit (CPU), or a dedicated processor for a specific purpose, such as a digital signal processor (DSP). In the foregoing, explicit use of the term "processor" should not be construed to refer exclusively to hardware capable of executing software, and may implicitly include, but is not limited to, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), Read-only memory (ROM), random-access memory (RAM), and non-volatile memory for storing software. Other hardware, conventional and/or custom, may also be included.

In the context of this specification, a "tag" is intended to mean a key-value tuple associated with each data center network device and stored in a database, such as a configuration management database (CMDB). A key uniquely identifies a data element and is either the actual data of the value or a pointer to the data, as the case may be. Labels are convenient keys to certain characteristics of data center networking equipment.

In the context of this specification, "BU" (for Business Unit) is intended, as part of the labeling of a data center network device, to denote a characteristic related to the business or product offering for which the data center network device is intended. The business or product offering may be that of the organization that deploys, controls and maintains the infrastructure or that of a customer of that organization. For example, a BU can be a server, cloud, host, etc., and depends on the granularity, variety, and complexity of the catalog of business or product offerings.

In the context of this specification, "ROLE" is intended, as part of the labeling of a data center network device, to denote characteristics related to the position and function that a data center network device occupies within the data center network infrastructure. For example, ROLE can be, but is not limited to: "aggregate", "top of rack" (ToR), "end of row" (EoR), "spine", "mega-spine", etc.

In the context of this specification, "INFRA" is intended, as part of the labeling of data center networking equipment, to denote the basis for which the data center networking equipment is intended to operate, which may evolve over time with enhancements and evolutions Version or generation related features of the facility.

In the context of this specification, "LOCATION" is intended, as part of the information associated with a data center networking device, to denote a characteristic relating to the actual physical location of a data center networking device in a data center. For example, LOCATION can be, but is not limited to: the name of a data center building, a specific data center room, and the like.

In the context of this specification, "MANAGEMENTIP" is intended, as part of the information associated with a data center network device, to denote a It performs operations such as eg retrieving information, changing configurations, upgrading etc. virtual identities such as eg IP addresses.

Still in the context of this specification, "a" computer readable medium and "the" computer readable medium should not be construed as the same computer readable medium. Conversely, "a" computer-readable medium and "the" computer-readable medium may also be construed as a first computer-readable medium and a second computer-readable medium, as appropriate.

Still in the context of this specification, unless expressly stated otherwise, the words "first", "second", "third", etc. have been used as adjectives only to allow the distinction between the nouns they modify distinction, not to describe any particular relationship between these nouns.

The embodiments of the present technology each have at least one of the above-mentioned objects and/or aspects, but not necessarily all of them. It should be understood that some aspects of the present technology resulting from an attempt to achieve the above purpose may not satisfy that purpose and/or may satisfy other objectives not specifically enumerated herein.

Additional and/or alternative features, aspects and advantages of embodiments of the technology will become apparent from the following description, drawings and appended claims.

Description of drawings

For a better understanding of the present technology, as well as its other aspects and further features, reference is made to the following description to be used in conjunction with the accompanying drawings, in which:

Figure 1 depicts a data center environment in which the present technology may be used;

Figure 2 presents a broad overview of methods according to the present technology;

Figures 3a to 3c provide more detailed views of methods in accordance with the present technology;

Figure 4 provides a logical illustration of clusters/sub-clusters of network devices in a data center; and

Figure 5 illustrates a computing system that may be used in the present technology.

It should be noted that unless otherwise expressly indicated herein, the drawings are not drawn to scale. Furthermore, elements that are the same from one figure to the next share the same reference label.

Detailed ways

The examples and conditional language enumerated herein are primarily intended to assist the reader in understanding the principles of the technology, not to limit its scope to these specifically enumerated examples and conditions. It will be appreciated that those skilled in the art can devise various arrangements that, although not explicitly described or shown herein, embody the principles of the technology and are included within its spirit and scope.

Furthermore, the following description may describe a relatively simplified implementation of the technology as an aid to understanding. Various embodiments of the present technology may have greater complexities, as will be appreciated by those skilled in the art.

In some cases, useful examples of what are considered modifications to the technology may also be set forth. This is done merely to aid understanding and, again, not to define the scope of the technology or to clarify the boundaries of the technology. These modifications are not an exhaustive list, and other modifications may be made by those skilled in the art while remaining within the scope of the technology. Furthermore, where no modified example is set forth, it should not be construed that no modification is possible and/or that what is described is the only way of implementing that element of the technology.

Moreover, all statements herein reciting principles, aspects, and embodiments of the technology, as well as specific examples thereof, are intended to encompass structural and functional equivalents thereof, whether currently known or developed in the future. Thus, for example, it will be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative circuitry embodying the principles of the technology. Similarly, it will be understood that any flowcharts, job diagrams, state transition diagrams, pseudocode, etc. represent various processes that may be substantially represented in a non-transitory computer-readable medium and thus executed by a computer or processor, whether explicitly Such a computer or processor is shown.

Software modules, or simple modules implied to be software, may be represented herein as any combination of flowchart elements or other elements indicating the execution of process steps and/or textual descriptions. Such modules may be performed by hardware shown explicitly or implicitly. Furthermore, it should be understood that a module may comprise, for example but not limited to, computer program logic, computer program instructions, software, stacks, firmware, hardware circuits or combinations thereof to provide the required capabilities.

With these rationales in place, we will now consider some non-limiting examples to illustrate various implementations of various aspects of the technology. Figure 1 depicts a data center environment in which the present technology can be used. Network upgrade tool 100 provides an input/output interface to data center operator 101 that allows the data center operator to enter inputs and options for automation of network equipment maintenance activities and to receive status and results regarding such activities . The network upgrade tool 100 can be developed using the YAQL language (Yet Another Query Language) and can be interpreted as a workflow within an open source software orchestrator for initiating and organizing tasks, subtasks and individual actions. For example, network upgrade tool 100 may be part of a software stack 102, such as Mistra, one of the components of the OpenStack project (available at https://docs.openstack.org/mistral/latest/). It will be apparent to those skilled in the art that other languages, software and software frameworks can still be used within the teachings of the present disclosure.

The network upgrade tool 100 may further interface with a network configurator 105 as an abstract and unified application programming interface (API) system that provides the ability to interact with network devices in the data center regardless of their hardware model or The diversity of network operating systems. The network configurator 105 may have an interface 106 with data center network devices.

Network upgrade tool 100 can further couple and interface with CMDB 103, which can reference network devices in the data center and store, for example, network upgrade tool 100 retrieved information associated with each of such referenced network devices. label and MANAGEMENTIP. For example, when each new network device is deployed in a data center, each new network device can be given a unique network device ID by, for example, the data center operator 101, and this network device ID can be used for retrieval and deployment from the CMDB 103 The label associated with the network device and MANAGEMENTIP.

The network upgrade tool 100 can further interface with an upgrade path DB 104, which can reference network devices in the data center and, as the case may be, each of such referenced network devices or such network devices The set of is associated with stores the path to the target network operating system level (ie: all necessary intermediate operating system levels between the current level and the target level) that the network upgrade tool 100 can retrieve and use to automatically maintain.

The network upgrade tool 100 can further interface with a rules DB 112, which can store rules that the network upgrade tool 100 can retrieve and use for automatic maintenance. For example, the network upgrade tool 100 can check whether certain network devices comply with certain rules. For example, a rule could detail the expected redundancy of network devices in a data center that share the same INFRA label value. For example, other rules might apply to network devices that share the same combination of BU-ROLE-INFRA tag values.

Those skilled in the art will appreciate that, although represented as three separate physical and logical entities, the CMDB 103, the upgrade path DB 104 and the rules DB 112 may all or partly form part of the same physical and/or logical database, or be located in a physical The above is part of the network upgrade tool 100 and does not affect the generality of the teachings herein. Furthermore, depending on the company and development environment of the organization deploying, controlling and maintaining the infrastructure, the rules DB 112 may not even be a separate database, and the rules may be hardcoded, for example in the network upgrade tool 100 .

A data center may include a plurality of network devices 109 interconnected by connections 110 in the infrastructure, as shown in FIG. 1 , as a simplified example only. Each network device 109 may have associated with it as part of its label:

-ROLE;

-INFRA 111; and

-BU 108.

In addition, LOCATION information 107a - 107b is also associated with each network device 109 .

Figure 2 presents a broad overview of methods in accordance with the present technology. In steps 201 and 202, certain parameters of the network devices 109 to be maintained in the data center (which may exist in the list of devices to be maintained) may be collected/obtained. This includes, at step 201 : one or more labels associated with each of the network devices 109 , including their BU, ROLE, INFRA and their MANAGEMENTIP and LOCATION information. And this includes, at step 202 : using the corresponding MANAGEMENTIP information of the network device 109 to obtain from the network devices certain parameters related to their hardware model and their current network operating system.

In step 203, those with unsupported hardware and software characteristics (such as unsupported hardware models or current network operating systems) that are not susceptible to the automatic maintenance of this technology or their BU, ROLE can be removed from the maintenance list. , INFRA label or those network devices 109 with wrong LOCATION information. For example, their network operating system level might not be a supported level, or their hardware model might be outdated or unrecognized. For example, their tags may not have a value recognized by the automated maintenance system of the present technology, or have (strictly) more or less than a single value for any of the BU, ROLE, or INFRA tags, etc.

At step 204, the remaining network devices 109 on the list may be clustered as follows:

- forming i clusters BU _i of network devices 109 sharing the same BU label;

- within each BU _i cluster, form j clusters ROLE _j of network devices 109 sharing the same ROLE label;

- Within each ROLE _j cluster, k clusters Cluster _ijk of network devices 109 sharing the same INFRA label are formed.

Those skilled in the art will appreciate that another sequence of cluster network devices 109 than described may be used to derive the configuration of the cluster Cluster _ijk , still within the teachings of this disclosure. For example, a cluster may first be composed of network devices 109 sharing the same ROLE label, then network devices 109 sharing the same BU label, and so on.

In step 205, in each formed Cluster _ijk , redundancy between network devices 109 may be identified. This redundancy may be defined at the INFRA _k level, with redundancy rules that automation systems of the present technology may apply to group redundant network devices 109 into groups within each Cluster _ijk . A group may include 1 (the network device has no redundancy) or 2 or more network devices 109 .

In step 206, in each formed group in the cluster Cluster _ijk , a specific size rule applicable to a specific combination of BU _i -ROLE _j -INFRA _k can be used to check the conformity of the network devices 109 present in the group . For example, the size rule may indicate that all groups in the cluster Cluster _ijk must include 3 network devices 109 . All network devices 109 belonging to a group comprising (strictly) more or less than 3 network devices 109 are non-compliant. Furthermore, a check may be performed whether all network devices 109 in the group share the same LOCATION information.

In step 207, those that were not clustered or grouped in steps 204 and 205, did not comply with redundancy rules or size rules, or did not share the same LOCATION information when present in the same group after steps 205 and 206 may be removed from the maintenance list Those network devices 109 .

At step 208, an automated process in accordance with the present technique may sequentially upgrade all remaining network devices 109 in the group. Because these network devices 109 are redundant network devices, interruptions to the functionality of the data center and to network connectivity services provided to customers of organizations deploying, controlling and maintaining the data center are minimized, and the data center Partial or total technical unavailability due to maintenance of network equipment is limited as much as possible. The same can be done for all created groups in Cluster _ijk and for all formed Cluster _ijk clusters. For tuned multiple Cluster _ijk clusters, the same operations can be performed in parallel to increase the speed of the maintenance process while minimizing the overall downtime of the data center and disruption of services to the customers of the organization that deploys, controls and maintains the data center . In addition, the order of processing Cluster _ijk clusters can also be adjusted to optimize the continuity of service to the organization's customers.

Figures 3a-3c provide more detailed views of methods in accordance with the present technology. Referring to FIG. 3 a , in step 301 , an attempt may be made to obtain information on the level of the current network operating system and the hardware model of the network device to be maintained in the data center. These network devices can be listed by maintaining a list of IDs on the list. Referring back to FIG. 1 , this can be performed, for example, by (i) providing (by data center operator 101 or otherwise) a maintenance list of network device IDs to network upgrade tool 100, (ii) network upgrade tool 100 collecting data from The corresponding MANAGEMENTIP of the CMDB 103, and (iii) the network upgrade tool 100 uses the retrieved corresponding MANAGEMENTIP to query the network device to be maintained to obtain the information of the current network operating system level and hardware model. The collection/acquisition is not necessarily successful, and the IDs of network devices whose collection/acquisition is unsuccessful can be added to the error list and removed from the maintenance list in steps 302 and 307.

In step 303, the remaining network devices on the maintenance list (that is: those network devices whose IDs are not added to the error list after steps 301/302) can be determined, whether the hardware model obtained in step 301 is supported hardware model. As used herein, "Supported Hardware Models" means the hardware models that Data Center's maintenance operations can handle. Referring back to FIG. 1, this can be performed, for example, by (i) storing supported hardware models (by the data center operator 101 or otherwise) in the rules DB 112, (ii) Retrieve those supported hardware models in , and (iii) the network upgrade tool 100 compares the retrieved supported hardware models with the hardware models obtained in step 301 . If the determination is unsuccessful or the hardware model is not supported, then in steps 304 and 307, the ID of the corresponding network device can be added to the error list and removed from the maintenance list.

In step 305, an attempt may be made to collect label information of the remaining network devices to be maintained in the data center on the maintenance list, as well as LOCATION and MANAGEMENTIP information. This can be performed by the network upgrade tool 100 collecting this information from the CMDB 103 . The collection is not necessarily successful, and in steps 306 and 307 the IDs of the network devices whose collection is unsuccessful can be added to the error list and removed from the maintenance list.

Referring to FIG. 3 b , in step 310 , it may be determined whether the remaining network devices on the maintenance list have one and only one value of the labels BU, ROLE and INFRA (as acquired in step 305 ). IDs of network devices with no value or more than one value in any of the BU, ROLE or INFRA tags may be added to the error list and removed from the maintenance list at steps 311 and 312 .

In step 313, the remaining network devices on the maintenance list may be grouped into clusters sharing the same BU label. IDs of network devices with unknown/unrecognized values in BU may be added to the error list and removed from the maintenance list at steps 314 and 312 .

In step 316, the remaining network devices on the maintenance list may be grouped into sub-clusters sharing the same ROLE label in each BU cluster. IDs of network devices with unknown/identified values in ROLE may be added to the error list and removed from the maintenance list at steps 317 and 312 .

Figure 4 provides a logical diagram of network devices in a cluster/sub-cluster data center according to steps 313 and 316 in Figure 3b. Only a portion of data center network equipment 401 is shown. The network devices 401 each have a unique ID which has been illustrated as "NDxx". As a result of step 313, network devices 401 are each clustered into a BU cluster 402, illustrated as BU01 and BU11. As a result of step 316, network devices 401 are each subclustered into ROLE subclusters 403, illustrated as ROLE01, ROLE07, ROLE13, and so on.

Returning now to Figure 3c, at step 320, for each cluster BU _i created at step 313 and each subcluster ROLE _j created at step 316, network devices can be further grouped into subgroups sharing the same INFRA label k (INFRA _k ) Subcluster Cluster _ijk .

In step 321, according to the redundancy rules applicable to the corresponding INFRA _k , the network devices that appear to be redundant in each Cluster _ijk may be further grouped into groups of redundant network devices. Referring back to FIG. 1 , this can be performed, for example, by (i) storing (by the data center operator 101 or otherwise) redundancy rules specific to INFRA _k in the rules DB 112, (ii) the network upgrade tool 100 Those redundancy rules are retrieved from the rules DB 112, and (iii) the network upgrade tool 100 groups network devices into groups in each Cluster _ijk according to those redundancy rules. For example, a rule may be the normalization of a network device ID that would inherently carry redundant information. For example, a rule could be that the IDs of redundant network devices in an INFRA _k end with the same chain of characters specific to that INFRA _k .

In step 322, it may be checked whether the hardware model of the network device existing in each group is one or one of several expected hardware models according to the rules applicable to the corresponding combination BU _i -ROLE _j -INFRA _k . Referring back to FIG. 1, this can be performed, for example, by (i) storing (by the data center operator 101 or otherwise) in the rules DB 112 hardware model expectation rules specific to BU _i -ROLE _j -INFRA _k , (ii) the network upgrade tool 100 retrieves those hardware model expected rules from the rule DB 112, and (iii) the network upgrade tool 100 checks the network devices in the groups in each Cluster _ijk according to those hardware model expected rules.

In step 323, the actual number of network devices present in each group may be checked according to the size rules applicable to the corresponding combination BU _i -ROLE _j -INFRA _k . Referring back to FIG. 1, this can be performed, for example, by (i) storing (by the data center operator 101 or otherwise) in the rules DB 112 specific sizes of BU _i -ROLE _j -INFRA _k (network device expected number) rules, (ii) the network upgrade tool 100 retrieves those size rules from the rule DB 112, and (iii) the network upgrade tool 100 checks the number of network devices in groups in each Cluster _ijk according to those size rules.

At step 324, it may be verified whether all network devices present in each group share the same LOCATION information. From step 305, the network upgrade tool 100 can use the LOCATION information.

In steps 325 and 326, may not be grouped in steps 320 or 321, did not match one of the expected hardware models in step 322, did not add up to the expected number in the group in step 323, or did not share when in the same group All network devices of the same LOCATION are added to the error list and removed from the maintenance list.

In step 327, the remaining network devices on the maintenance list may be maintained: for all network devices in the group, for all Cluster _ijk , this is performed sequentially. Referring back to FIG. 1, this can be performed, for example, by (i) storing (by the data center operator 101 or otherwise) in the upgrade path DB 104 the OS level upgrade rules, (ii) the network upgrade tool 100 retrieves those upgrade rules from the upgrade path DB 104 and (iii) the network upgrade tool 100 applies the upgrade through the network configurator 105 .

With this technology, the network devices involved in providing the same service to the same customers in the data center have been identified and grouped together, and after removing those network devices that are not compliant and for which automated maintenance is too risky, the All network devices in the group perform maintenance sequentially. Therefore, the service is not completely disrupted, but degraded as appropriate.

Those skilled in the art will appreciate that the granularity of the error lists referred to in steps 307, 312 and 326 can be higher by tracking the cause of the errors. For example, the list of errors may be broken down into sublists that allow errors such as labeling, classification, identification, and (ii) linking to faulty network devices such as through their network operating system level Distinguish between errors that are linked to faulty network devices, etc.

Figure 5 illustrates a computing system that may be used in the present technology. An example of an implementation of a computing system 500 that may be used with network upgrade tool 100 and/or network configurator 105 is presented. As will be understood by those skilled in the art, such a computing system may be implemented in any other suitable hardware, software, and/or firmware, or a combination thereof, and may be a single physical entity or several separate physical entities with distributed functionality .

In some aspects of the present technology, computing system 500 may include various hardware components including one or more single-core or multi-core processors collectively represented by processor 501 , solid-state drive 502 , memory 503 , and input/output interface 504 . In this context, processor 501 may or may not be included in an FPGA. In some other aspects, computing system 500 can be an "off-the-shelf" general-purpose computing system. In some aspects, computing system 500 may also be distributed among multiple systems. Computing system 500 may also be specific to embodiments of the present technology. As can be appreciated by those skilled in the art, numerous changes may be envisaged as to how computing system 500 may be implemented without departing from the scope of the present technology.

Communications between the various components of computing system 500 may be through one or more internal and/or external buses 505 (e.g., PCI bus, Universal Serial Bus, IEEE 1394 "FireWire" bus) to which the various hardware components are electronically coupled. , SCSI bus, Serial ATA bus, ARINC bus, etc.) to achieve.

Input/output interface 504 may allow for enabling networking capabilities, such as wired or wireless access. As an example, input/output interface 504 may include a networking interface such as, but not limited to, a network port, network socket, network interface controller, and the like. Several examples of how the networking interface may be implemented will become apparent to those skilled in the art. Solid state drive 502 may store program instructions, such as those portions of libraries, applications, etc., adapted to be loaded into memory 503 and executed by processor 501 for methods and process steps in accordance with the present technology, according to embodiments of the present technology.

Although the foregoing embodiments have been described and illustrated with reference to particular steps performed in a particular order, it will be understood that these steps may be combined, subdivided, or reordered without departing from the teachings of the disclosure. At least some of the steps may be performed in parallel or serially. Accordingly, the order and grouping of steps is not a limitation of the present technology. It should also be clearly understood that not all technical effects mentioned herein need to be enjoyed in every embodiment of the technology.

Modifications and improvements to the above-described embodiments of the technology may be apparent to those skilled in the art. The foregoing description is intended to be illustrative rather than limiting. Accordingly, the scope of the technology is intended to be limited only by the scope of the appended claims.

Claims (15)

1. A method for maintaining a plurality of network devices in a data center, comprising:

collecting, for at least one network device of the plurality of network devices, the at least one network device selected from a maintenance list:

-a management IP, the management IP being a virtual identity uniquely associated with the at least one network device;

-a location, which is information representative of an actual physical location of the at least one network device; and

-a tag being a set of key values associated with the at least one network device, the key values being one of data and pointers to the data,

the tag is one of:

-a BU tag, said BU tag being information representative of a product provisioning application of said at least one network device in said data center;

-a ROLE tag being information representative of a function occupied by the at least one network device in the data center; and

-an INFRA tag, the INFRA tag being information representative of a version of the data center in which the at least one network device is operating;

for the at least one network device on the maintenance list, acquiring a hardware model and a current network operating system level of the at least one network device based on the management IP of the at least one network device; and

in response to determining at least one of:

the hardware model of the at least one network device is not supported,

the current network operating system level of the at least one network device is not supported, an

An error in at least one of the BU tag, the ROLE tag, the INFRA tag, and the location tag,

removing the at least one network device from the maintained list.

2. The method of claim 1, further comprising determining at least one of:

the hardware model of the at least one network device is not supported,

the current network operating system level of the at least one network device is not supported, an

An error in at least one of the BU tag, the ROLE tag, the INFRA tag, and the location tag.

3. The method of claim 1, further comprising:

collecting, for each network device of the plurality of network devices:

the management IP is a set of IP addresses that are managed,

said position, and

the label;

in response to removing the at least one network device from the maintenance list, clustering remaining network devices on the maintenance list into:

i clusters of BUs of network devices having the same BU tag value among the associated tags of the remaining network devices _i ，

Nets having the same ROLE key value in one of the associated labels of the remaining network devicesNetwork device j cluster ROLE _j And ROLE of a network device having the same INFRA tag value in one of the associated tags of the remaining network devices _j K clusters within a Cluster _ijk ；

At each Cluster _ijk Creating a redundant group of the plurality of network devices within the cluster; and

according to the suitability for INFRA _k Verifying whether the number of redundant network devices in each created group matches the first number; and

according to applicable combination BU _i -ROLE _j -INFRA _k Verifying whether the number of network devices in each group matches the second number.

4. The method of claim 3, further comprising:

verifying whether the network devices of each group share the same location;

removing from the maintenance list the following network devices: at least one network device present in a given group and having a number of redundancies that does not match the first number; at least one network device present in another given group in a number that does not match the second number; or at least one network device present in another group and not sharing the same location.

5. The method of claim 4, further comprising:

all Clusters are managed using the upgrade rules and corresponding management IPs _ijk The remaining network devices in each group within the cluster upgrade from a current version of the network operating system to a target version of the operating system.

6. The method of claim 1, wherein the collecting and the obtaining are performed using a unique ID assigned to each of the network devices on the maintenance list by a data center operator at a time of deployment of the data center, and the unique ID is associated with any of the associated tags of each network device, the management IP, and the location.

7. The method of claim 5, wherein removing from the maintained list further comprises: network devices that do not have exactly one value for each of the BU tag, the roll tag, and the INFRA tag are removed.

8. The method of claim 6, wherein removing from the maintained list further comprises: an error list is created that is populated with the IDs of the removed network devices.

9. The method of claim 8, wherein the error list further comprises: a sub-list of errors that can be corrected by the data center operator and errors that link to a failed network device.

10. The method of claim 6, wherein the upgrading further comprises: adjusting Cluster for parallel processing _ijk A number of clusters to minimize downtime of the data center.

11. The method of claim 5, wherein the upgrading further comprises: adjusting process Cluster _ijk An order of clustering to optimize continuity of service to customers of an organization that deploys, controls, and maintains the data center.

12. A system for maintaining network devices in a data center, the system comprising a network upgrade tool coupled to a CMDB, a rules DB, an upgrade path DB, and a network configurator, the network upgrade tool configured to:

-collecting from the CMDB for each of a plurality of network devices on a maintenance list:

-a management IP, said management IP being a virtual identity uniquely associated with a given network device;

-a location, which is information representative of an actual physical location of the given network device; and

-a label being a key-value tuple associated with the given network device,

wherein the tag is one of:

-information, BU, representing a service or product offering for which the network device is used in the data center;

-information representative of the locations and functions occupied by said network devices in said data centre, ROLE; and

-information representative of a version or generation of the data center, INFRA, in which the network device operates;

the value for the tag is the actual data or a pointer to the data;

-for each of the network devices on the maintenance list, obtaining the hardware model and current network operating system level of the given network device using the respective management IP;

-removing from the maintenance list: a network device having an unsupported hardware model; or a network device with a current network operating system level that is not supported; or a network device having an error in any of its respective associated BU tag, roll tag, and INFRA tag or in any of its locations;

-clustering the remaining network devices on the maintenance list into: i clusters BU of network devices having the same BU key value in one of the associated labels of the remaining network devices _i (ii) a In BU _i J cluster ROLEs of network devices having the same ROLE key value in one of the associated labels of the remaining network devices within a cluster _j (ii) a And ROLE of network devices having the same INFRA key in one of the associated tags of the remaining network devices _j K clusters within a Cluster _ijk ；

-adapting the INFRA according to what is collected from the rule DB _k At each Cluster _ijk Creating redundant network devices within a clusterPreparing the group;

according to the suitability for use in INFRA _k Verifying whether the number of redundant network devices in each group matches the first number;

according to applicable combination BU _i -ROLE _j -INFRA _k Verifying whether the number of network devices in each group matches the second number;

-verifying whether all network devices in each created group share the same location;

-removing from the maintenance list the following network devices: network devices present in one of the groups and having a number of redundancies that does not match the first number; a number of network devices present in one of the groups that does not match the second number; or network devices that are present in one of the groups and that do not share the same location; and

-using the upgrade rules and corresponding management IPs collected from the upgrade path DB, by the network configurator, for all clusters _ijk All groups within the cluster upgrade the remaining network devices in each group from a current version of the network operating system to a target version of the operating system.

13. The system of claim 12, wherein the network upgrade tool is further configured to: the collecting and retrieving is performed using a unique ID assigned to each of the plurality of network devices on the maintenance list by a data center operator at the time of deployment of the data center, and the unique ID is associated with any of the associated tags of each network device, the management IP, and the location.

14. The system of claim 12, wherein the redundancy rules and the size rules are hard-coded in the network upgrade tool.

15. A computer-readable medium comprising instructions that cause a computing system to perform the method of claim 1.

Images